Silicon Wafer Oxide Thickness Prediction via Furnace Aging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Accurately predicting the thickness of an oxide layer on a silicon wafer is challenging due to the narrow tolerance range and variations in growth conditions, which affects the uniformity and quality of semiconductor devices.
Innovation Solution
A method involving the use of a heat treatment furnace to form oxide layers on reference and test wafers with different pretreatment conditions, measuring the thickness at multiple points, and alternating their placement to determine factors affecting the oxide layer growth, including surface roughness and cleaning conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Length of moving object
If the oxide layer thickness is reduced to achieve thinner layers, then the tolerance range becomes smaller and measurement precision requirements increase, but manufacturing precision becomes more difficult to maintain
Solution Approach 1:
The furnace is aged before actual oxide layer formation by processing reference wafers first. This preliminary action stabilizes the furnace environment and oxidation conditions, ensuring that subsequent test wafers experience consistent conditions that enable precise thickness control even at reduced thicknesses
Solution Approach 2:
The method systematically varies oxidation parameters (temperature, time, atmosphere) and measures resulting oxide layer thicknesses to establish predictive relationships. By understanding how parameter changes affect thickness, the process can precisely control oxide layer formation at reduced thicknesses while maintaining uniformity
2Measurement precision
If multiple measurement points are taken to improve measurement accuracy, then measurement precision increases, but the complexity of the measurement process increases
Solution Approach 1:
The measurement process is segmented into two phases: first measuring reference wafers to establish baseline furnace performance and oxidation characteristics, then using this data to predict and validate measurements on test wafers. This segmentation reduces overall complexity while maintaining precision through systematic data collection
3Manufacturing precision
If reference wafers and test wafers are processed separately to maintain control, then manufacturing precision is maintained, but productivity decreases
Solution Approach 1:
Reference wafers and test wafers are merged into the same heat treatment boat and processed simultaneously in the same furnace under identical conditions. This combining approach maintains precision by ensuring consistent treatment while improving productivity by processing multiple wafers in parallel rather than sequentially
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method allows for precise prediction and control of oxide layer thickness, reducing variations and ensuring high uniformity and purity, essential for advanced semiconductor devices like epitaxial wafers.
Implementation Method 1
forming oxide layers
Data Source
AI summary
An embodiment provides a method of predicting a thickness of an oxide layer of a silicon wafer including: aging a heat treatment furnace (furnace); measuring a thickness of each of the oxide layers after disposing a plurality of reference wafers in slots of a heat treatment boat in the furnace and forming oxide layers; and measuring a thickness of each of the oxide layers after disposing the plurality of reference wafers and test wafers in the slots of the heat treatment boat in the furnace and forming oxide layers.


